[12:06] <fabbione> i can't believe that in the last 3 weeks everything, and i mean everything, couldn't go worst
[12:07] <npmccallum> mdz: good, I've been working with thom on the initscript stuff today
[12:07] <npmccallum> mdz: I'll have another batch in tonight
[12:07] <npmccallum> mdz: with this batch it *should* be all of warty main
[12:35] <lamont__> am I really here?
[12:35] <fabbione> yup
[12:35] <lamont__> 159kB/s. give or take
[12:35] <lamont__> 20 minutes more of sitting in the street outside someone's house.
[12:36] <lamont__> and I'm smarter now, in that I know the 70W inverter just doesn't cut it when the vaio is running flat out...
[12:36] <lamont__> but then, 3*19 should be < 70 W...
[12:36] <lamont__> well, 19.5
[12:40] <lamont__> found a part of the CD that hadn't changed...
[12:40] <lamont__> 2MB/s
[12:44] <fabbione> ehhe
[12:45] <fabbione> i have 2MB/s here anyway :-))
[12:46] <lamont__> fabbione: megabits, or bytes?
[12:47] <fabbione> bits
[12:47] <lamont__> the burst I got was bytes.. :-)
[12:47] <fabbione> ok something here is seriously fucked
[12:48] <lamont__> fabbione: you working on X again??? :-)
[12:48] <fabbione> i have been building this package all the day and it was working
[12:48] <fabbione> now it doesn't compile anymore
[12:48] <fabbione> lamont__: hehehe no
[12:48] <lamont__> no updated depends or anything?
[12:48] <fabbione> i really don't understand wtf is wrong
[12:48] <fabbione> lamont__ it's the damn nvidia driver
[12:49] <fabbione> it was building fine until 1 hour ago
[12:49] <fabbione> when the raid on my server died
[12:49] <fabbione> Mithrandir:
[12:49] <fabbione> 1) the server does not have the source
[12:49] <fabbione> ops
[12:49] <fabbione> s/Mithrandir/ The server/
[12:49] <fabbione> 2) it doesn't build
[12:50] <lamont__> although when you're bored, anything you can tell me to make an ATI Radeon 7500 or ATI Rage Mobility P/M go faster than stock-warty would be cool.
[12:50] <Mithrandir> fabbione replaced me with a server
[12:50] <fabbione> 3) i only logged out and logged in on my workstation
[12:50] <fabbione> AH
[12:50] <fabbione> and i didn't even change the debian/rules or anything
[12:50] <fabbione> that's the funny part
[12:51] <lamont__> so it's the environment, then?
[12:53] <fabbione> it didn't change...
[12:53] <fabbione> Your kernel was configured to include rivafb support!
[12:53] <fabbione> IT DOESN
[12:53] <fabbione> IT DOESN'T!
[12:53] <fabbione> grrrrrrrrrrrrrrrrrrrrrr
[12:54] <fabbione> before this check was passing
[12:54] <fabbione> and it gets that info from -headers
[12:57] <fabbione> AH AH
[12:57] <fabbione> got it
[12:58] <mdz> fabbione: what was it?
[01:02] <lamont__> mdz: yet another initrd-tools uploaded
[01:03] <fabbione> mdz: something that was not cleaned/reinstalled correctly
[01:06] <fabbione> mdz: i also left something for you in the source package
[01:06] <mdz> fabbione: for me? you shouldn't have
[01:06] <fabbione> mdz: README-for-security-teams_or_to-update-the-nvidia-driver
[01:07] <mdz> but it's not my birthday yet!
[01:07] <fabbione> mdz: ehheheh
[01:07] <fabbione> mdz: just a few notes of what to check for security updates or driver update
[01:07] <fabbione> the nvidia stuff is NOT simple and linear
[01:08] <mdz> README.NMU seems to be an emerging standard for documentation like that
[01:08] <fabbione> linux-restricted-modules-2.6.8.1-2.6.8.1 -> linux-restricted-modules-2.6.8.1-2.6.8.1.0
[01:08] <mdz> though that name is perhaps not applicable to ubuntu :-)
[01:08] <fabbione> as source version
[01:08] <mdz> fine with me
[01:08] <fabbione> is that ok for you?
[01:08] <mdz> or .1
[01:08] <mdz> or +nvidiaX
[01:08] <mdz> or whatever
[01:08] <fabbione> .0
[01:09] <fabbione> :-)
[01:09] <mdz> it probably shouldn't have 2.6.8.1 in it at all; it doesn't correspond to the kernel really
[01:09] <mdz> not in the source package
[01:09] <mdz> but 2.6.8.1.0 is more than OK with me
[01:11] <lucas_> hi
[01:14] <seb128> hey lucas_ 
[01:22] <lamont__> Kamion: you awake still?
[01:23] <thom> FUCKING DAN JACOBSON
[01:24] <fabbione> dan... dan...
[01:24] <fabbione> it's not a new name
[01:24] <Clint> jidanni
[01:25] <thom> fabbione: see #1148-1150
[01:26] <thom> http://bugs.debian.org/cgi-bin/pkgreport.cgi?submitter=jidanni@jidanni.org
[01:29] <fabbione> AHAHAHAHAHA
[01:29] <fabbione> i want his crack!
[01:29] <fabbione> i want his crack!
[01:29] <fabbione> i want his crack!
[01:29] <thom> too right
[01:30] <mdz> he just tried to bump this silly, ancient mirror bug up to severity: critical
[01:30] <mdz> this guy did not get enough attention as a child
[01:30] <Clint> at least he's not lazarus long
[01:31] <mdz> how do you know?
[01:31] <Clint> if he is, then I will be suitably impressed.
[01:32] <mdz> :-)
[01:32] <mdz> he is at least as bad
[01:32] <Clint> less hostile
[01:32] <mdz> I remember lazarus long filing pointless bugs, but not going around and screwing up bug severities
[01:32] <Clint> though I didn't really expect him to look like http://jidanni.org/me/images/20040321dan.jpg
[01:32] <Clint> no, lazarus long would frequently file minor things as critical and be very adamant about the severity
[01:33] <lucas_> Computer -> System configuration -> Login screen setup -> "Failed to run gdmsetup as user root: Child terminated with 1 status". Somebody can reproduce ?
[01:33] <mdz> Clint: dan jacobson repeatedly fucks with the severity after somebody fixes it, though
[01:33] <mdz> lucas_: works for me; did you enter the correct password?
[01:33] <Clint> well, he's not wrong 100% of the time
[01:34] <mdz> he's close enough
[01:34] <mdz> I am completely fed up with him filing bugs against apt
[01:34] <lucas_> mdz, yes, I double checked
[01:34] <mdz> lucas_: your password, and not the root password?
[01:34] <fabbione> mdz: you got an email :-)
[01:34] <fabbione> and i am off to sleep
[01:34] <lucas_> mdz, sorry for this =)
[01:34] <Clint> heh, I've gotten private warning mails about his bugs
[01:35] <lucas_> bbiaf
[01:36] <thom> lazarus is certainly more irritating
[01:36] <mdz> but he went away, didn't he?
[01:36] <mdz> or did I just killfile him?
[01:37] <mdz> I don't seem to have, so apparently he went away
[01:37] <thom> yeah, he seems to come and go 
[01:37] <mdz> I don't think he ever filed the quantity of bugs that dan jacobson has, either
[01:38] <thom> oh sure. but jacobson has never refused to help debug problems afaik
[01:38] <mdz> he's in the top 5 of all time
[01:38] <mdz> he certainly has
[01:38] <mdz> almost all of the bugs I have received from him are not even bugs, so he can't exactly help debug them
[01:39] <mdz> he will file a severity: grave bug against apt saying I THINK THIS COMMAND LINE OPTION SHOULD BE DIFFERENT
[01:39] <fabbione> ahahha
[01:39] <Clint> either he went away or changed his name
[01:40] <mdz> he was also the one who went around filing bugs about whitespace
[01:40] <mdz> whitespace!
[01:40] <mdz> he said it slowed his computer down
[01:40] <mdz> or wasted space or something
[01:40] <thom> hahah
[01:40] <mdz> I shit you not
[01:40] <thom> that's awesome
[01:40] <mdz> that was not the word I used for it at the time
[01:41] <thom> no, i imagine not
[01:42] <thom> i think lazarus long's irssi "no, i won't tell you what additonal modules i have loaded to cause irssi to crash" was the all time winner
[01:42] <mdz> if you check my recent debian bugs or usenet postings, you'll see that
[01:42] <mdz> i have caught many 'whitespace slobs' e.g. fetchmail who adds up to
[01:42] <mdz> two for each line in syslog... also i suggested this as a lintian or
[01:42] <mdz> deb policy reminder item. anyways, i'm pointing it out now as who
[01:42] <mdz> knows, next month i might turn my (setq-default
[01:42] <mdz> show-trailing-whitespace t) emacs x-ray specs off and then be as blind
[01:42] <mdz> as everybody else to this minor, but still wasteful area in creating
[01:42] <mdz> the common goal of a sleek fine tuned debian.
[01:43] <mdz> http://jidanni.org/ Taiwan(04)25854780
[01:43] <mdz> is that a telephone number?
[01:43] <thom> some people have *far* too much time on their hands
[01:43] <thom> yeah, looks like it should be
[01:43] <mdz> I didn't notice that at the time
[01:44] <Clint> or you would have called him?
[01:44] <thom> mdz: probably a good thing
[01:44] <elmo> dan's never replied to any of my replies to his bugs
[01:44] <mdz> I don't know if I would have _called_ him, per se
[01:45] <mdz> he might have started to receive more phone calls than he might have otherwise
[01:45] <thom> *g*
[01:45] <Clint> I actually haven't found him to be too annoying when I haven't ignored him.
[01:46] <Clint> or maybe I've blocked that out
[01:46] <mdz> maybe we should calling you Job rather than Clint
[01:46] <mdz> s/calling/start &/
[01:47] <Clint> I certainly don't have a knee-jerk fuming anger reaction when I get a bug from him.
[01:47] <mdz> I really tried to help him at first, I did
[01:47] <Clint> which I can't say about everyone
[01:47] <mdz> I pointed him to all sorts of good documentation about how to write good bug reports
[01:47] <mdz> since he seemed to _want_ to help
[01:47] <mdz> but he ignored everything I told him
[01:47] <mdz> repeatedly
[01:48] <mdz> it seems that when he gets bored, he sits down and writes a few hundred bug reports in batch mode, and then dials up his modem and uploads them
[01:49] <mdz> he never checks whether his bug has been filed, because he would need to dial up
[01:49] <Clint> doesn't he work for an ISP?
[01:49] <mdz> given that he complains about his internet access constantly, I didn't think so
[01:49] <mdz> he filed a bunch of bugs about packages being too big for his modem
[01:50] <Clint> I must be thinking of some other twit in .tw
[01:50] <mdz> asking that things be split up in the most useless ways so that he could avoid downloading another 10k
[01:51] <Clint> hmm, I should do something about #213700
[01:53] <elmo> nah, Dan's on a modem, there's no way he works for an ISP
[01:54] <Clint> elmo: yeah, somehow I merged him with Jason Lim in my head
[01:54] <mdz> he does occasionally stumble on a real bug from time to time, apparently through sheer luck
[01:54] <mdz> but chances are it's already been reported anyway
[01:55] <elmo> mdz: I think he once admitted to not checking the BTS before filing because it took up too much BW
[01:55] <mdz> elmo: yes, he has admitted that he never does
[01:55] <mdz> that's also his excuse for not reading about how to use the BTS
[01:55] <Clint> what the hell speed modem is he on?
[01:57] <elmo> something slow enough that each pixel of whitespace that's rendered causes him offence
[02:04] <thom> i'm surprised he's not filed a bug about bugs.d.o not doing gzip Content-Encoding
[02:06] <Clint> thom: email him immediately to suggest it
[02:07] <elmo> oh, i finally just read d-d to see what you mean.. yeesh
[02:09] <thom> Clint: tempting, but Colin knows where I live
[02:09] <thom> he'll come and kill me
[02:10] <Clint> tricky
[02:16] <Kamion> elmo: seeds fixed
[02:16] <Kamion> (mea culpa, etc.)
[02:16] <Kamion> lamont__: ?
[02:17] <Kamion> if Dan argues, he goes in the control@bugs sin-bin
[02:18] <elmo> dude, aj would never euphemize @gFuckHeads as 'control@bugs sin-bin' :)
[02:20] <Kamion> heh
[02:21] <Kamion> I wish he had, then people wouldn't believe it did other things ...
[02:21] <Kamion> I did make a start on general debbugs access control recently, to stop people abusing sarge-ignore I think
[02:22] <elmo> cool
[02:23] <thom> night folk
[02:23] <lamont__> Kamion: question answered
[02:23] <elmo> night thom
[02:47] <Kamion> what's the python idiom for "make a shallow copy of this sequence"?
[02:48] <spiv> list(seq)
[02:49] <Kamion> ah yes, I knew I'd seen it before, thanks
[02:50] <Kamion> the documentation mentions seq[:]  (the sequence in this case happens to be a list); is that considered obscure?
[02:51] <spiv> Well, it's a known idiom, but I don't find it as readable.
[02:53] <spiv> I mean, if you understand slicing syntax it's clear what it does (and has the potential advatage that it'll return the same type of sequence, rather than casting to a list)... but it's just looks odd to my eye.
[02:53] <spiv> s/casting/converting/
[02:54] <Kamion> spiv: right, thought so
[03:05] <Kamion> elmo: is the lvmcfg sync request still in your queue? sorry, I sent the request while the package was still in Debian incoming
[03:08] <elmo> kamion: gar, sorry, no, it's just I've been redoing my mail filtering based on canonical.com stuff and it went to a folder I don't usually check for syncs ... done now
[03:10] <Kamion> ta
[03:20] <dieman> lamont__: you just need to get an ubertv and then you can avoid that
[03:20] <dieman> :)
[03:20] <lamont__> dieman: Bourne Supremacy is not on TV yet...
[03:21] <dieman> ahh
[03:21] <dieman> yeah
[03:21] <dieman> there is that
[03:21] <dieman> we just do netflix
[03:21] <dieman> ive got the same tv bdale has
[03:21] <dieman> found his glowing comments on it from his website and was sold
[03:21] <dieman> we were trying to figure out wich one to get, but when the first link back for the tv is a DD, how can you argue?
[03:22] <dieman> (or one of the first)
[03:38] <spiv> Whee... I'm seeing "Error activating XKB configuration" even with the version that bug 1089 claims fixes it.
[03:47] <mdz> lamont: gone yet?
[03:53] <lamont> mdz: just got home, will be leaving in about 20-30
[04:00] <lamont> who claims linux-restricted-modules-2.6.8.1_2.6.8.1.0-1
[04:01] <elmo> dude, it's already been fixed
[04:01] <lamont> even better
[04:01] <elmo> linux-restricted-modules-2.6.8.1 | 2.6.8.1.0-2 | warty/restricted | source
[04:02] <lamont> nvidia-settings is ftbfs on ppc...
[04:02] <lamont> or rather, 1.0-3 was.
[04:03] <elmo> night all
[04:04] <lamont> g'night elmo
[04:13] <mdz> lifeless: around?
[04:14] <lamont> mdz: why were you looking for me, I wonder???
[04:14] <mdz> lamont: linux-restricted-modules and friends
[04:16] <lamont> mdz: ah, ok.
[04:17] <lamont> mdz: could it be that growisofs is broken differently?
[04:18] <lamont> growisofs -Z /dev/hdc foo
[04:18] <lamont> :-( unable to INQUIRY: Bad file descriptor
[04:19] <mdz> lamont: it's been working perfectly for me today
[04:19] <mdz> shouldn't that be growisofs -Z /dev/hdc=foo?
[04:19] <lamont> if foo was an ISO..
[04:20] <mdz> oh, it's a directory?
[04:20] <mdz> dunno
[04:20] <mdz> but I've burned 4 or 5 things today with growisofs under warty and haven't encountered that problem
[04:20] <mdz> but all ISOs
[04:20] <lamont> 5.19.4.9.7-2ubuntu2
[04:20] <lamont> ??
[04:20] <mdz> lifeless: ipw2200 0.7 isn't working for me, while 0.4 did
[04:21] <mdz> lamont: the one where I fixed the open mode
[04:21] <mdz> rebooting that box right now
[04:22] <mdz> Version: 5.19.4.9.7-2ubuntu2
[04:23] <spiv> mdz: My gf has had lots of trouble with her ipw2200 on her new laptop.  She needs to reboot on average once an hour to get it to work.  Seems like the ipw2200 driver is still rather rough :(
[04:24] <lamont> open("/dev/hdc", O_RDONLY|O_LARGEFILE)  = 4
[04:24] <lamont> fstat64(4, {st_mode=S_IFBLK|0660, st_rdev=makedev(22, 0), ...}) = 0
[04:24] <lamont> open("/dev/hdc", O_RDWR|O_LARGEFILE)    = -1 EROFS (Read-only file system)
[04:24] <lamont> ioctl(-1, 0x2285, 0xbffff768)           = -1 EBADF (Bad file descriptor)
[04:24] <lamont> write(2, ":-( unable to INQUIRY: ", 23:-( unable to INQUIRY: ) = 23
[04:24] <lamont> write(2, "Bad file descriptor\n", 20Bad file descriptor
[04:24] <lamont> looks like you missed one...
[04:24] <mdz> that one's all yours
[04:24] <lamont> yeah - I'm working on it.
[04:24] <mdz> I've had enough GOTO for the week
[04:25] <mdz> there are two definitions of that function, but the other one seems to be #ifdef'd for BSD
[04:26] <lamont> oh, gag.
[04:26] <lamont> what if I just make all of the opens RDWR?
[04:26] <lamont> 1/4 :-)
[04:27] <lamont> off to the movies - will deal with it when I get back.
[05:27] <daniels> #(*@$IOJDF HEISENBUG
[05:28] <npmccallum> lol
[05:31] <mdz> daniels: have you done a fresh Warty install recently?
[05:39] <daniels> mdz: yes, when i got home
[05:40] <daniels> roughly a fortnight ago
[05:40] <mdz> daniels: that's not recently :-)
[05:40] <mdz> daniels: I think doing a fresh install would be a good step in debugging this
[05:40] <mdz> maybe you'll run into the problem too
[05:40] <daniels> um
[05:40] <daniels> a) unless we get jigdo working for my local mirror, downloading the iso is non-trivial for me (involves being somewhere else, really)
[05:41] <daniels> b) would you like me to reinstall my development machine?
[05:41] <daniels> i've added debugging to xresprobe, and the version of X i hope to upload today will have debugging also
[05:41] <mdz> a) you were supposed to have your bandwidth situation sorted out a long time ago, and yet you were still spending time on it yesterday
[05:41] <mdz> b) everyone needs to be doing installation testing
[05:41] <daniels> i can rsync and upload, that's fine
[05:41] <daniels> i have that aspect sorted out
[05:42] <daniels> but if i suddenly need to be pulling 600MB and have it done quickly, I need to find someone who's at home and bum off their DSL
[05:42] <daniels> given both of my housemates are out today and I thus have no car
[05:42] <daniels> in general, it's fine, but I can't cop 650MB spikes on no notice
[05:42] <daniels> and I plan to gain another machine with my next paypacket -> installer testing
[05:43] <mdz> look, we have a public release in 5 days
[06:35] <cef> daniels: I should be able to download you a copy of the iso and drop it around
[06:36] <cef> daniels: and I can give short term a celeron on an i815 m/board with a 6 gig drive
[06:36] <daniels> cef: dude, that would be absolutely sensational, thanks
[06:36] <cef> mdz: which image does he need?
[06:36] <daniels> cef: sounder 8
[06:37] <cef> I've got sounder8 burnt, but if you need a daily, can be done
[06:37] <daniels> nah, sounder 8 is fine
[06:37] <cef> cool.. already downloaded then
[06:37] <daniels> rad
[06:38] <daniels> i should be home all day, so whenever's fine for you
[06:38] <cef> cool. I've got to do some washing shortly, so I can drop past then.
[06:39] <daniels> sure, that'd be great, thanks :)
[06:39] <cef> do you need the celeron too? it's not assembled at the mo, but that's no biggie
[06:39] <daniels> that would really help, yeah
[06:40] <cef> ok.. only got 128 meg of ram in it tho.. but should be enuff for a test methinks... soon to be my router *grin*
[06:40] <cef> bbk
[06:41] <daniels> heh :)
[06:41] <daniels> fo'sho
[06:41] <cef> tis 1/2 height pci
[06:41] <cef> so nice and small.. always good for a router
[06:42] <daniels> heh, nice
[06:42] <daniels> yeah
[06:42] <daniels> i've never really had small pcs
[06:42] <cef> btw: if you need not-so-hot AGP video cards of weird descriptions (eg: voodoo3's and the like).. ask
[06:42] <cef> cos we have a whole heap of shit at work we're throwing shortly.. and the cards are fine, just not really used anymore
[06:44] <cef> thinking they might be useful for someone working with X.. *grin*
[06:44] <daniels> cool, thanks, yeah ... if you're chucking any ont, more cards is always a good thing
[06:44] <daniels> especially, nvidia, much as I dislike them; all I have right now is my radeon 8500, and a 9000 at home
[06:45] <daniels> both r200
[06:45] <cef> no probs.. they're all at work, so I'll bring em home monday
[06:46] <daniels> awesome, cheers
[06:46] <cef> cool.. got a few tnt's, tnt2's, possibly a 440mx, an old pci matrox millenium, a few voodoo 3's, possibly a voodoo4, lots of S3 Trio64's, and some ATI mach64 chipset cards. and that's just off the top of my head
[06:47] <daniels> wow
[06:47] <daniels> could i please grab a tnt2, a 440mx if you have it, the milennium, a voodoo3, a trio64, and a mach64?
[06:47] <daniels> :)
[06:47] <cef> and of course, as we're moving, we've gotta throw all this shit out, so... *grin*
[06:47] <cef> I'll justy grab all the video cards and you can have the lot. *grin*
[06:49] <daniels> heh! awesome, cheers :)
[06:49] <daniels> lemme know if there's anything you need
[06:49] <cef> no probs
[06:49] <daniels> although i've mostly been cleaned out ... pretty much all i have left is a usb nic, a volcano 9 ...
[06:50] <cef> heh
[06:58] <cef> couldn't find the 6gig drive.. so you'll have to live with 10
[06:58] <cef> *grin*
[07:05] <daniels> heh :) wot pain!
[07:06] <cef> just make sure the cdrom will read the bootdisk
[07:08] <cef> ouch just hit a possible bug
[07:09] <cef> twas using the ubuntu installer just going to departition the disk.. it hanging during detecting hardware.. hrm.. could be an ide cable problem.. bbk
[07:09] <daniels> 5hm
[07:09] <daniels> anything from dmes,g or was it a solid lock?
[07:10] <cef> solid lock.. I've had this beforfe with ide cables, and this one isn't in the best of condition
[07:10] <cef> and I've got like 4 spare *grin*
[07:10] <daniels> heh :)
[07:11] <daniels> you should put it in bz - computers suck
[07:11] <cef> moved my desktop to round ide cables, so all these spare flat ones are hanging about
[07:13] <daniels> ah, heh
[07:13] <daniels> i got some rounded ones - rally did cut down on the clutter
[07:14] <cef> same
[07:14] <cef> ahh looks much better now
[07:15] <cef> and yeah, broken/damaged wires in the ide cable
[07:15] <daniels> heh, in the rounded cable?
[07:16] <cef> nah, a flat one that was in this machine I'm loaning you
[07:17] <daniels> ah, rad
[07:17] <daniels> broken rounded cables wouldn't be so cool -- when i got mine, they were like 15 or 20 bucks a pop for halfway decent ones
[07:18] <cef> yeah..
[07:41] <cef> just having something to eat, then I'll head over
[07:45] <lifeless> mdz: yo
[07:45] <lifeless> still here?
[07:46] <daniels> cef: cool, ta
[07:46] <daniels> lifeless: it's 0200
[07:47] <lifeless> I don't know these things :}
[07:47] <lifeless> besides hes cto of the distro: how can he have time to sleep just now
[07:47] <cef> yeah, but given the type of people we have here, you have to always check *grin*
[07:52] <cef> hrm, do we have any ppl from alsa-devel in the ubuntu team, pref within australia?
[07:53] <kagou> hi
[07:53] <daniels> cef: nope
[07:53] <cef> just realised I've got a few spare sound cards as well that are looking for a good home.. *grin*
[07:53] <daniels> cef: heh, yeah. given our team meeting the other day started at 0200, my bedtimes have been 0340, 0220, 0200 and 0557 the past couple of days
[07:53] <daniels> cef: heh :)
[07:57] <cef> btw: once I get this machine in place as a router, I'll have the current machine free.. Athlon 500 on an amd/via combo chipset board. if you're interested, you're welcome to it once it's decommisisoned
[07:57] <cef> got too much hardware here.. need to get rid of some
[07:58] <cef> got 2 machines sitting here for colin charles actually.. he needs to get off his read and come pick em up
[07:59] <cef> and we'l be chucking out some more machines at work soon
[08:00] <lamont> mdz still around?
[08:03] <daniels> oooer
[08:04] <daniels> does the via board have a via video chipset?
[08:09] <cef> nah.. currently got a nvidia tnt in it.. no on board.. if I see anything obscure like that, I'll pass it your way
[08:10] <lamont>                 if (fp->f_mode & FMODE_WRITE) {
[08:10] <lamont>                         ret = -EROFS;
[08:10] <lamont> sigh.
[08:13] <cef> daniels: ok, cya in 10
[08:18] <daniels> cef: rad, thanks :)
[09:05] <fabbione> morning guys
[09:07] <lamont> night fabbione
[09:07] <fabbione> night lamont
[09:09] <AndyFitz> http://216.239.39.104/translate_c?hl=en&u=http://perso.wanadoo.fr/kitone/sabrina/site/features.html&prev=/search%3Fq%3Dlinuxgraphic.org%26hl%3Den%26lr%3D%26ie%3DUTF-8
[09:09] <AndyFitz> /me drools
[09:09] <AndyFitz> night lamont
[09:28] <mdz> lamont: here now
[09:58] <cef> daniels: how'd it go?
[10:03] <jdub> yo alextreme 
[10:03] <alextreme> gmorning :)
[10:03] <alextreme> cool, hope it works for you :)
[10:26] <daniels> cef: yeah, worked well, thanks :)
[10:27] <daniels> couldn't find any of these bugs
[11:04] <kagou> hi
[11:04] <kagou> mdz, is that you managing pppoeconf ?
[11:21] <debianist> morning folks
[11:35] <kagou> hello debianist
[11:36] <debianist> hi kagou
[11:37] <kagou> i'v problem with pppoeconf... do i post a bug report on "ppp" ?
[11:37] <debianist> nvidia drivers are in restricted already? can I try and install them?
[11:37] <kagou> *i don't knox
[11:37] <kagou> i don't know
[11:37] <debianist> kagou : try look for pppoeconf
[11:37] <Kamion> kagou: if the component doesn't exist, please use UNKNOWN.
[11:37] <debianist> yes,
[11:37] <debianist> and that's also :)
[11:37] <kagou> ok Kamion
[11:38] <debianist> morning Kamion
[11:38] <Kamion> (rather than picking a similarly-named package at random)
[11:38] <Kamion> morning
[11:38] <Kamion> the nvidia drivers went into restricted last night
[11:38] <debianist> Kamion : cool
[11:38] <debianist> Kamion : any how on the wiki for installing? (I reckon it'd be good idead for the new users to have one there)
[11:38] <Mithrandir> moo
[11:39] <debianist> moo Mithrandir
[11:39] <Kamion> debianist: not on the public wiki yet; it's part of the preparation for the preview-release launch
[11:41] <kagou> ati drivers are in process ?
[11:41] <debianist> Kamion : kool , i'll try install them now :) 
[11:53] <Mithrandir> I have a weird problem.. when I burn a cd, my USB controller freaks out:
[11:53] <Mithrandir> uhci_hcd 0000:00:10.1: host controller halted, very bad!
[11:53] <Mithrandir> uhci_hcd 0000:00:10.2: host system error, PCI problems?
[11:53] <Mithrandir> uhci_hcd 0000:00:10.2: host controller process error, something bad happened!
[11:54] <Mithrandir> a reboot doesn't fix it, I have to power off
[01:08] <pitti> Hi guys!
[01:10] <debianist> hey pitti
[01:28] <Keybuk> "I just moved into the area and the aircraft noise is TERRIBLE!  My children can't SLEEP at night!"
[01:28] <Keybuk> ... perhaps you shouldn't have moved into the village at the end of the runway then?
[01:29] <debianist> :)
[01:45] <pitti> doko: there?
[01:51] <doko> pitte: yes
[01:52] <doko> pitti: yes
[01:52] <pitti> doko: I just read about your DSA script
[01:52] <pitti> does it filter out the non-relevant packages?
[01:52] <pitti> or even the version number where it got fixed?
[01:53] <doko> both, I could sort out 70% of the DSA for 2003
[01:53] <pitti> doko: debianist is processing the 2002 DSAs right now, I think he would welcome it
[01:53] <pitti> doko: if that saves to much time, *GIVE IT TO ME* :-))
[01:53] <doko> ok, I send it.
[01:54] <pitti> doko: that's great! thanks
[01:54] <pitti> doko: debianist has a new mail address now
[01:54] <pitti> debianist: please give doko your new mail
[01:54] <debianist> doko : sivang@gmail.com
[02:03] <doko> pitti, debianist: mail sent.
[02:03] <debianist> doko : thanks
[02:07] <pitti> doko: thanks!
[02:09] <pitti> doko: I did not yet get it. Where did you sent it to?
[02:11] <debianist> doko : me nither..
[02:12] <doko> ok, I sent it from another address.
[02:14] <debianist> doko : got it
[02:15] <pitti> doko: got it, thanks a lot!
[02:17] <lifeless> daniels: around?
[02:18] <doko> lifeless: while you are around: where should libsvncpp{0,-dev} go?
[02:18] <lifeless> doko: hell.
[02:18] <lifeless> they are eveeeeeel
[02:19] <doko> ?
[02:19] <lifeless> what sort of answer were you expecting ?
[02:21] <doko> well, you wanted them, you got them, only comment: "eve^6l"
[02:21] <lifeless> so, are you asking "which suite they should go in" 
[02:21] <lifeless> or which dir?
[02:21] <lifeless> or ... ?
[02:26] <doko> yes, which suite ...
[02:29] <lifeless> doko: ok, erm. for warty it probably should be universe
[02:30] <lifeless> for debian it should be optional
[02:31] <doko> ok, fine.
[02:33] <lifeless> cool.. thank you
[02:33] <lifeless> I wasn't meaning to be unappreciative.. it was just my 'unqualified question, silly answer' reflex
[02:38] <doko> I think I got used to it on the conference, even without questions ;)
[03:01] <daniels> lifeless: um, sort of
[03:01] <daniels> lifeless: if you need me, call me
[03:06] <lifeless> nothing major
[03:06] <lifeless> just get corrupt screen shots, was wondering about causes.
[03:07] <lifeless> only menus are problematic
[04:02] <debianist> how can I know the original package a ubuntu one is based on? how can I see which patches has been applied on the move from the sid package, to the ubuntu one?
[04:09] <Kamion> debianist: look at the version number and drop ubuntu*; download the two source packages, run debdiff (from the devscripts package)
[04:10] <debianist> Kamion : thank you :)
[04:18] <daniels> lifeless: screen shots, or actual display? using which version of x?
[04:18] <daniels> (any more questions and I'll be helping you debug tla problems or some shit)
[04:19] <daniels> does your doctor's surname have a z in the third or seventh positions? ;)
[04:20] <HrdwrBoB> if so, you should seek help immeadietly
[04:20] <Kamion> ... from a different doctor
[04:21] <HrdwrBoB> yes!
[04:22] <daniels> it would also explain why your pristine inode revlib caches are corrupt
[04:33] <Kamion> hm, at least I think that's what it is ...
[04:44] <daniels> oh, duh
[04:45] <daniels> spiv: try s/hpi6/inet(hpi6)/ ?
[04:46] <Kamion> I think we need a bit more support in netcfg to get prism2_usb devices going, given that they apparently don't support the standard wireless extensions interface
[04:48] <daniels> Kamion: where are the base modules set? i'm trying to fix #1139, but neither d-i nore base-config have any mention of psmouse
[04:49] <Kamion> can you detect whether psmouse is needed?
[04:49] <Kamion> if not, is it harmful to load it on systems that don't have a PS/2 mouse?
[04:50] <Mithrandir> shouldn't be
[04:51] <daniels> Kamion: with mdetect, but then if you plug one in later, you don't get it loaded, and no
[04:51] <daniels> Kamion: i can't see why it shouldn't be loaded unconditionally
[04:51] <Kamion> then do that in ddetect
[04:51] <Kamion> there are similar examples there already
[04:52] <Kamion> basically run 'register-module psmouse' somewhere appropriate
[04:52] <Kamion> if you could avoid doing it for powermacs I'd appreciate it; there's already an if arch = powerpc there
[04:54] <spiv> daniels: Just did; see my update to the bug :)
[05:02] <doko> how do I enter i bug report for a non fixed DSA in ubuntu, which is in universe, but not in warty? I can't enter the component ...
[05:02] <Mithrandir> doko: UNKNOWN
[05:02] <HrdwrBoB> put UNKNOWN
[05:02] <doko> thanks, ugh, bugzilla already cleared out the form :-(((
[05:12] <HrdwrBoB> hm
[05:12] <HrdwrBoB> I think having gamma in gnome-display-properties would be nice
[06:06] <daniels> spiv: cheers
[06:06] <daniels> Kamion: rad, thanks
[06:08] <daniels> Kamion: just to make sure i'm not totally on crack, is the i386 $SUBARCH, i386(?
[06:08] <daniels> er, i386?
[06:08] <daniels> er
[06:08] <daniels> i386*
[06:08] <Kamion> i386/*, and you probably want amd64/* too
[06:09] <Kamion> I assume you're looking in get_input_info
[06:09] <Kamion> oh, and don't echo "psmouse:whatever"; that's for when you want the module loaded in d-i too
[06:09] <daniels> Kamion: cheers
[06:09] <daniels> right
[06:10] <Kamion> if you worked that out, congratulations, you're smarter than Sven Luther ... ;-)
[06:32] <spiv> daniels: Have you seen thedailywtf.com?
[06:34] <Clint> wow, Kamion with the sanity on -policy
[06:48] <Kamion> my god, you and Thomas B *both* think it makes sense?
[06:48] <Kamion> the world must be ending
[06:54] <Clint> I didn't say it made sense; I said you brought sanity.
[06:58] <Kamion> Clint: :-)
[06:59] <doko> the 20040910 test build works like a net-installer CD, about 550MB size, but installs only about 400MB (unpacked) and starts downloading things after the reboot... anyone knows what is going wrong?
[07:00] <Clint> and now I get to remove 11 more operators from posh's "test"
[07:00] <Kamion> doko: updates from ftp.no-name-yet.com? it adds that to sources.list
[07:00] <Kamion> doko: the CD contains stuff (the Ship seed) that isn't installed by default
[07:01] <doko> ok, but that's the current CD to test? no X installed/configured?
[07:04] <Kamion> uh, it should definitely install X ...
[07:09] <doko> is there a newer cd for download?
[07:10] <Kamion> 20040911 will be the latest
[07:11] <Kamion> not significantly different from 20040910 that I'm aware of, though
[07:12] <doko> ok, there is 11-Sep-2004 08:25, I'll try that one.
[07:14] <Kamion> I'm just booting it now
[07:16] <Clint> does prism2_usb not work with hostap?
[07:20] <Kamion> I can give it a try
[07:20] <Kamion> there's no mention of USB in ieee80211.dpatch though
[07:21] <Kamion> hm, no, that's not the hostap code though
[07:21] <Kamion> do we actually have hostap in our kernels?
[07:22] <Kamion> <cjwatson@riva ~/src/canonical/linux-source-2.6.8.1/linux-source-2.6.8.1-2.6.8.1>$ find -name \*hostap\*
[07:22] <Clint> no, you don't
[07:22] <Kamion> <cjwatson@riva ~/src/canonical/linux-source-2.6.8.1/linux-source-2.6.8.1-2.6.8.1>$
[07:22] <Kamion> tricky to try it in d-i then :)
[07:22] <Clint> I just thought it would integrate better than wlan-ng
[07:22] <Kamion> might well do
[07:22] <Kamion> http://lists.shmoo.com/pipermail/hostap/2004-March/006076.html
[07:23] <Kamion> seems not
[07:23] <Clint> ah well
[07:30] <doko> kamion: should the installer ask for a proxy before trying to start the download during the installation?
[07:35] <Kamion> yes, please file a bug on base-conifg
[07:35] <Kamion> base-config
[07:36] <Kamion> fabbione's changes were on a different code path and so totally bypassed the proxy code
[07:39] <doko> kamion: is there a question in this path as well, which "mirror" to use for the downloads (i.e use my configured apt-proxy)?
[07:40] <Kamion> nope
[07:41] <doko> hmm, reinstalling now, I think I know what I did differently during the first install: I did break the download, edited my sources.list to use the aptproxy, and tried to restart. but what if the connection breaks and it's not a break of your choice?
[07:43] <Kamion> that should drop you into aptitude, shouldn't it?
[07:44] <doko> yes
[07:44] <Kamion> if that happens, install the ubuntu-desktop task
[07:44] <Kamion> we'll document this in the installation manual ...
[07:45] <doko> ok, I did reinstall and I get to the desktop
[07:49] <mdz> morning folks
[07:50] <doko> good evening ;)
[07:51] <npmccallum> good afternoon
[07:51] <doko> npmccallum: east coast?
[07:54] <npmccallum> doko: yeah :)
[07:59] <doko> mdz, npmccallum: concerning the ISDN support, I'll continue the work tonight and tomorrow, but I'm now uncertain that we will get something useful for warty, so I would like to concentrate to get the components into universe in such a way, that they are available and usable. however, it would be nice to prepare the init scripts (currently isdnutils-basse, isdnactivecards, maybe misdn not yet available) to follow the conventions for init script
[07:59] <doko> s. would it be possible that npmccallum cleans up these scripts?
[08:00] <mdz> doko: there is much to do for the Wednesday preview release; I think we should revisit this issue after then
[08:00] <mdz> if we are not overwhelmed by the feedback from the preview, we can consider this
[08:00] <mdz> otherwise, we can make it a goal for Hoary
[08:01] <doko> thats ok, so the goal for warty (IMO) should be to have the packages available in universe, whithout configuration assistance.
[08:06] <mdz> daniels, Kamion: is 1.03ubuntu5 the right solution to the problem?
[08:06] <mdz> (ddetect)
[08:06] <Kamion> mdz: works for me
[08:06] <mdz> ok
[08:07] <Kamion> if it isn't autodetectable and loading it is harmless, then ...
[08:07] <mdz> but why now?
[08:07] <mdz> it autoloads fine for me on i386, and I haven't seen any reports to the contrary
[08:07] <mdz> I guess i am asking what problem it is solving
[08:08] <mdz> Steve McIntyre reported that it was not autoloaded for him on amd64
[08:10] <mdz> thom: ping?
[08:10] <Mithrandir> is "it" psmouse?
[08:12] <Kamion> hotplug appears to know how to load psmouse under some circumstances?
[08:12] <Kamion> it's in the isapnp code
[08:12] <Kamion> I can well believe that isapnp doesn't work on amd64
[08:15] <mdz> Mithrandir: yes
[08:15] <mdz> psmouse uses the serio hooks to get autoloaded
[08:15] <mdz> I really don't know how that works
[08:17] <mdz> but presumably something on amd64 should notice there is a serial port and call that stuff
[08:27] <kagou> hi guys
[09:32] <debianist> mdz : i have a package which according to bugtrail have been patched and there
[09:32] <debianist> mdz : 's even a web spot to download the patched one 
[09:32] <Mithrandir> mdz: ooo-amd64 is uploaded now, btw.  It probably needs a bit of NEW love, though
[09:32] <debianist> mdz : however i cannot find it on p.d.o, debian havn't updated it yet?
[09:33] <mdz> debianist: can you be more specific?
[09:33] <mdz> Mithrandir: it probably needs 6 hours to build, too :-)
[09:33] <Mithrandir> mdz: nope, builds in about 30 secs.
[09:34] <Mithrandir> mdz: as you have to prepare the package on i386
[09:34] <mdz> Mithrandir: hmmm?
[09:34] <mdz> Mithrandir: does it use the binaries built for the i386 deb or something?
[09:34] <Mithrandir> yes.
[09:34] <Mithrandir> but it ships the sources with the .debs
[09:34] <Mithrandir> so we're ok in that regard.
[09:34] <Mithrandir> fucking ugly
[09:35] <Mithrandir> (ok, license-wise, I mesan)
[09:35] <Mithrandir> s/mesan/mean/
[09:35] <mdz> so -m32 was not sufficient?
[09:35] <mdz> what do we do when we need to do a security update on oo.o, what is the procedure?
[09:35] <debianist> mdz : ofcourse. bugzilla 2.16.6-1_all.deb the patched one, available under http://www.tzi.de/~jmm/debian/ . however sid has  2.16.5-2: all
[09:35] <debianist>  - warty is using that.
[09:36] <Mithrandir> debuild on i386, put the sources in srcs/, the i386 binaries in pkgs/, build on amd64.
[09:36] <Mithrandir> that is, debuild ooo on i386
[09:36] <Mithrandir> it's documented in debian/README.build
[09:38] <debianist> mdz : this is #260773
[09:39] <Kamion> hm, current installations are supposed to pop up a window on the desktop when I insert a USB stick, right?
[09:40] <debianist> it did for me :)
[09:41] <Kamion> shows up in /dev, is pmountable, but doesn't automount ...
[09:41] <debianist> Kamion : however I had problem umounting it. couldn't find it's mount point, and nothing on the desktop to handle it :(
[09:42] <Kamion> whoa, what happened to hald?
[09:44] <Kamion> it's segfaulting, great
[09:47] <daniels> spiv: heh, no
[09:54] <mdz> Kamion: that would explain it
[09:54] <mdz> debianist: I'll look at the bug, one moment
[09:54] <Mithrandir> mdz: you seemed unhappy about OOO, or?
[09:55] <mdz> Mithrandir: it is obviously not ideal, but as long as the procedure is documented
[09:55] <mdz> Mithrandir: then I can live with it for Warty
[09:55] <debianist> mdz : thanks.
[09:55] <mdz> Mithrandir: so is it actually built from a different source package on amd64?
[09:55] <Mithrandir> mdz: yes
[09:56] <Mithrandir> mdz: I'm not happy with it either, but I don't think implementing multiarch $RSN is a real option.
[09:56] <mdz> Mithrandir: we don't need multiarch, we just need ooo to build and work on amd64 :-)
[09:56] <Mithrandir> yeah, and we want gcc to be able to cross-compile stuff.
[09:57] <mdz> debianist: the patch is here: http://bugzilla.mozilla.org/show_bug.cgi?id=235265
[09:57] <mdz> debianist: it is straightforward and we can use that patch in warty
[09:57] <Mithrandir> it's probably less work to implement multiarch than fixing OOO. :)
[09:58] <mdz> ooo will get fixed on its own, maybe for hoary if we are lucky
[09:58] <debianist> mdz : ok, so tag it fixed?
[09:58] <mdz> debianist: no?
[09:58] <mdz> we have the bug in warty and need to fix it before we can tag it fixed :-)
[09:58] <mdz> likewise for Debian
[09:58] <debianist> mdz : ofcourse
[09:59] <debianist> mdz : but what about the "fixed" pacakges by MOritz?
[09:59] <mdz> debianist: file a bug in Warty bugzilla with the URL of the patch
[09:59] <mdz> debianist: there is no reason to bring in some unofficial packages with a new upstream release to fix this simple issue; we can patch it
[09:59] <debianist> mdz : so it become standard debian?
[09:59] <mdz> debianist: public release in _four days_ :-)
[10:00] <debianist> mdz : i know mdz, i'll open a bug and continue the overnight rush :)
[10:00] <mdz> debianist: I don't understand your question about standard debian
[10:00] <debianist> mdz : nevermind. I just didn't understand why we cannot check the unofficial package and make it official debian
[10:01] <debianist> mdz : if the job has already been done
[10:01] <mdz> debianist: there are two things here
[10:01] <mdz> debianist: one is a new upstream release, with the fix _and_ other changes
[10:01] <mdz> debianist: the other is a patch which only has the fix
[10:01] <debianist> mdz : ok
[10:02] <mdz> debianist: hmm
[10:02] <mdz> debianist: according to the changelog there are actually three bugs, not one
[10:03] <mdz> CAN-2004-0704, CAN-2004-0705, CAN-2004-0706
[10:05] <mdz> in this case perhaps we should take the new upstream
[10:05] <mdz> justdave: around?
[10:06] <debianist> mdz : you're looking at the CAN site?
[10:09] <mdz> debianist: I looked at the changelog for the unofficial package
[10:09] <mdz> there is also CAN-2004-0707
[10:10] <mdz> SQL injection vulnerability in editusers.cgi in Bugzilla 2.16.x before 2.16.6, and 2.18 before 2.18rc1, allows remote attackers with privileges to grant membership to any group to execute arbitrary SQL.
[10:10] <mdz> debianist: have you filed a bug yet?  if not, I will do it so that I can record these findings
[10:11] <debianist> mdz : doing it right now, would you like me to reference it to the other bugs?
[10:11] <mdz> debianist: just open the bug, so that I can add comments to it
[10:11] <debianist> mdz : sure.
[10:23] <mdz> debianist: what's the bug number?
[10:24] <debianist> mdz : darn. bugzilla just sent me back due to "bad component name" i'll re enter
[10:25] <debianist> mdz : we don't bugzilla on the components list
[10:25] <mdz> debianist: use UNKNOWN
[10:28] <debianist> mdz : #1161
[10:28] <mdz> debianist: thanks
[10:29] <debianist> mdz : sorry about the UNKNOWN thingy, I just went and started typing, not noticing bugzilla was not identified..
[10:34] <debianist> mdz : the patch addresses all 4 CANs?
[10:34] <mdz> debianist: no
[10:35] <mdz> there are three
[10:35] <mdz> and they are all addressed in 2.16.6
[10:35] <mdz> and also with patches in upstream bugzilla
[10:35] <mdz> CAN-2004-0706 does not apply to 2.16.x, only 2.17.x
[10:35] <debianist> k
[10:40] <justdave> mdz: am now
[10:40] <mdz> justdave: wanted to ask about CAN-2004-0704, CAN-2004-0705, and CAN-2004-0707 (now #1161 in warty bugzilla)
[10:40] <mdz> justdave: is 2.16.6 a safe bet for a near-to-release update to Warty?
[10:41] <justdave> yeah, that has all those addressed.
[10:41] <justdave> Rmi hasn't put one up yet?
[10:43] <justdave> hmm, nope, looks like he hasn't.
[10:44] <justdave> unless he backported the fixes to 2.16.5 (which would be silly since that's the only change between it and 2.16.6 anyway :)
[10:45] <mdz> justdave: remi seems to be on vacation
[10:45] <mdz> justdave: see http://bugs.debian.org/235265
[10:46] <justdave> kdemultimedia?
[10:46] <justdave> that the right bug number?
[10:47] <mdz> er
[10:47] <justdave> 260772, 260773, 260774 I see
[10:47] <mdz>  #260773
[10:47] <mdz> yes
[10:48] <mdz> 235265 is the bugzilla.mozilla.org bug number for one of them
[10:48] <mdz> justdave: I confirmed that 2.16.6 fixes the bugs, but how much of a risk are we taking with other changes?
[10:49] <mdz> since 2.16.6 isn't in Debian yet, it has probably seen relatively little testing in Debian environments
[10:49] <justdave> not much, that's the stable branch, it doesn't get touched except for security and low-risk stuff...
[10:50] <justdave> but it depends on how much of the fixes hit spots near stuff he's patched
[10:50] <justdave> 2.18 isn't as bad to debianize, but 2.16.x had to be hacked up pretty bad to make it fit in Debian
[10:52] <justdave> when I grab a deb-src, does it have the patch applied already on the directory it unpacks?
[10:53] <justdave> bah, no patches, just the big diff
[10:59] <justdave> the debian patch applies to a 2.16.6 tarball with no conflicts
[11:01] <justdave> +  * Include complete copyright statements collected from the sources
[11:01] <justdave> +    (Closes: #253841)
[11:09] <debianist> justdave : you patched the bugzil pkg?
[11:10] <justdave> I did locally, I haven't put it anywhere yet.  I'm comparing it to the one Moritz posted right now
[11:11] <debianist> justdave : great. let me know when  you have it verified.
[11:15] <justdave> ok, only differences between mine and his are the addition of that copyright file and a permissions fix in the cgi-bin directory in the postinst, which isn't mentioned in his changelog
[11:17] <debianist> mdz : if he uploads the new package we can close #1611 and i'll remove it from the wiki
[11:17] <justdave> mdz: what's permissions policy for the cgi-bin directory?
[11:17] <debianist> justdave : see #1611 that i filed and mdz changed.
[11:18] <justdave> world-executable okay or should it be group executable to www-data only?
[11:18] <debianist> i'd say group executable to www-data, but we better ask mdz about this.
[11:19] <justdave> Moritz's added line to the postinst makes it world executable
[11:19] <mdz> justdave: yes, it applies all the patches when it unpacks it
[11:20] <mdz> justdave: yes, I saw that postinst change and I don't understand what it's for
[11:20] <mdz> justdave: I don't think that system() should be there
[11:21] <mdz> the files should be in the .deb with proper permissions
[11:22] <justdave> hmm, he has the /var/cache/bugzilla permissions fix in there twice, too.
[11:23] <justdave> the unpacked tarball appears to have everything with owner-only execute permissions
[11:23] <justdave> but it looks to be chowning everything to www-data:www-data anyway
[11:24] <mdz> the CGIs should obviously not be owned by www-data
[11:24] <mdz> I hope he hasn't done that
[11:25] <justdave> fix_www_data_perm('/var/lib/bugzilla');   #this should be done by checksetup.pl
[11:25] <justdave> fix_www_data_perm('/var/cache/bugzilla'); #but I dislike the way this is done.
[11:25] <justdave> sub fix_www_data_perm {
[11:25] <justdave>         my $path = shift;
[11:25] <justdave>         system(qq{chown -R www-data.www-data $path}) == 0
[11:25] <mdz> hmm
[11:25] <mdz> bugzilla is not in warty/main
[11:25] <justdave> that's one ugly postinst
[11:26] <justdave> this is actually the first time I've taken a good look at the debian directory in the bugzilla.deb
[11:27] <mdz> the package has always been a bit of a mess, because as you say, it was difficult to map bugzilla to FHS and debian policy
[11:27] <mdz> but you say it's easier with the new releases, so hopefully it'll clean up
[11:29] <justdave> what checksetup.pl does is a lot cleaner/more secure than what he's doing in the postinst (which actually runs checksetup.pl first), but he's got that entire section commented out in checksetup.pl
[11:30] <justdave> probably would have worked better to fix the pathnames in checksetup.pl and let it do it.
[11:30] <justdave> it chowns the user doing the install, and chgrps to www-data (group name is configurable) with it only accessible to the group
[11:39] <justdave> how long will Debian support woody with security fixes after sarge is released?
[11:45] <mdz> in theory, 6 months or so
[11:47] <mdz> debianist: the changelog in the unofficial 2.16.6 package
[11:47] <debianist> mdz : k
[11:47] <mdz> debianist: also http://bugs.debian.org/cgi-bin/pkgreport.cgi?src=bugzilla
[11:50] <debianist> mdz : k,tnx again
[11:57] <justdave> Rmi has a 2.17.7 deb up on alioth
[11:58] <justdave> that's prior to the advisory though, so some of the same bugs still
[11:58] <justdave> (not to mention, that's the development branch)